Modeling sub-surface jets and assessing water quality implications

摘要

Hydropower accounts for 65.9% of renewable electricity generation and 7% of total electricity generation in the USA. This provides a safe, reliable, efficient and cost effective renewable energy source. Downstream water quality is one of the major challenges faced by hydropower industry. Elevated Total Dissolved Gas (TDG) concentrations have been observed downstream of large hydropower facilities which can be detrimental for fish and ecology. In order to achieve compliance to Federal and State water quality regulations (regarding TDG) during re-licensing, existing hydropower facilities have to modify dam operation and/or retrofit the spillways to avoid high TDG concentrations downstream. This is an expensive, time-consuming and uncertain process. The purpose of this research is to make this process easier, less expensive and more predictable. One approach is to use diversion tunnels and other structures from the construction phase to reduce spillway flow. These tunnels usually release water close to or below the downstream water surface. With sufficient velocity, the jet can drag down the water surface, entrain air and increase TDG concentrations. We are investigating the physical processes involved in the development of sub-surface jets. A Computational Fluid Dynamics (CFD) model of sub-surface jet flow is developed using open-source CFD code OpenFOAM. We validated the model by simulating an experimental setup and comparing the results with measurements. The model uses Reynolds Averaged Navier Stokes (RANS) turbulence model and Volume of Fluid (VOF) method for water surface tracking. The validated CFD model will be expanded to full scale to model the Cabinet Gorge Hydroelectric Project, Idaho. The dam has existing tunnels which can be used to divert spillway flow. The CFD model will simulate the flow from the tunnels into the plunge pool and will be used to investigate the effect of flow rate, jet velocity, and release depth/tail-water depth on air entrainment and jet development. The findings of this research will provide a better understanding of sub-surface jet flow and eliminate uncertainties in the design/decision making process. This paper illustrates how standard RANS (k-ε), k-ω, and RNG (renormalization group) k-ε models do not represent measured data while a modified RANS k-ε model was successful.